Core Concepts
SigNova framework improves RFI detection in radio astronomy data.
Abstract
The article introduces the SigNova framework for detecting anomalies in streamed data, focusing on radio-frequency interference (RFI) in radio astronomy. The framework comprises three primary components: signature transform for feature extraction, novelty score calculation using Mahalanobis distance, and integration with Pysegments for anomaly localization. SigNova outperforms SSINS and AOFLAGGER in identifying various types of RFI, even localizing faint RFI effectively.
Introduction to Radio Astronomy and RFI challenges.
SigNova framework overview: signature transform, novelty score calculation, and Pysegments integration.
Comparison of results with SSINS and AOFLAGGER using simulated and real data.
Performance analysis of SigNova in detecting RFI across different datasets.
Stats
SigNovaは、新しいストリーミングデータの異常検出フレームワークです。
Mahalanobis距離を使用して各特徴ベクトルに異常スコアを割り当てます。
SigNovaは、SSINSおよびAOFLAGGERよりもさまざまな種類のRFIを検出することができます。